Cannabinoids have been reported to sustain self-administration in laboratory animals; however, genetic differences and environmental factors critical in the initiation and retention of such behaviour are yet to be defined. This study investigated the acquisition, maintenance and extinction of self-administration of the cannabinoid CB1 receptor agonist WIN 55,212-2 (6.25-25 microg/kg/inf) in Long Evans, Lister Hooded and Sprague-Dawley rats under a continuous schedule of reinforcement and two different response-like operanda, nose-poking and lever-pressing. Results showed that Long Evans and Lister Hooded, but not Sprague Dawley, rats acquired and retained stable cannabinoid self-administration behaviour under both modus operandi, as defined by significant differences between responding in the active versus the inactive hole/lever. In rats developing firm self-administration, substitution of saline for WIN 55,212-2 extinguished the responding, supporting the notion that cannabinoids may serve as a positive reinforcer in laboratory animals. Nevertheless, significant differences among strains and responding modalities were observed in the percentage of acquisition, amount of drug intake during maintenance and timing of extinction. In addition, no significant strain differences were found in motor response to WIN 55,212-2 (0.3 and 3.0 mg/kg), thus excluding that strain differences observed during cannabinoid self-administration could be related to different cannabinoid-induced locomotor effects.